Costimulatory Function of Cd58/Cd2 Interaction in Adaptive Humoral Immunity in a Zebrafish Model

Abstract

CD58 and CD2 have long been known as a pair of reciprocal adhesion molecules involved in the immune modulations of CD8⁺ T and NK-mediated cellular immunity in humans and several other mammals. However, the functional roles of CD58 and CD2 in CD4⁺ T-mediated adaptive humoral immunity remain poorly defined. Moreover, the current functional observations of CD58 and CD2 were mainly acquired from in vitro assays, and in vivo investigation is greatly limited due to the absence of a Cd58 homology in murine models. In this study, we identified cd58 and cd2 homologs from the model species zebrafish (Danio rerio). These two molecules share conserved structural features to their mammalian counterparts. Functionally, cd58 and cd2 were significantly upregulated on antigen-presenting cells and Cd4⁺ T cells upon antigen stimulation. Blockade or knockdown of Cd58 and Cd2 dramatically impaired the activation of antigen-specific Cd4⁺ T and mIgM⁺ B cells, followed by the inhibition of antibody production and host defense against bacterial infections. These results indicate that CD58/CD2 interaction was required for the full activation of CD4⁺ T-mediated adaptive humoral immunity. The interaction of Cd58 with Cd2 was confirmed by co-immunoprecipitation and functional competitive assays by introducing a soluble Cd2 protein. This study highlights a new costimulatory mechanism underlying the regulatory network of adaptive immunity and makes zebrafish an attractive model organism for the investigation of CD58/CD2-mediated immunology and disorders. It also provides a cross-species understanding of the evolutionary history of costimulatory signals from fish to mammals as a whole.

Keywords: Cd4+ T cells; adaptive humoral immunity; cd2; cd58; costimulatory signals; zebrafish.

Figure 1

(A,B) The cDNA sequences and deduced amino acids of cd58 and cd2. The signal peptide and transmembrane domain are underlined with the dotted and solid lines, respectively. Two conserved cysteines in Cd58 and four in Cd2 amino acid sequences forming IgC2-like domain structures are encircled, respectively. The asterisk represents the stop codon. (C,D) Comparison of the intron/exon organizations of the CD58 and CD2 genes in humans and zebrafish. Exons and introns are shown with the black boxes and lines, and their sizes are indicated by the numbers found above and below the sequences, respectively. Black and white areas indicate the coding regions and untranslated regions, respectively. Schematic diagrams were included above the exon/intron organization cartoons to indicate the exons in terms of protein domains, including signal peptide, IgV domain, IgC2-like domain, transmembrane domain (TM), and cytoplasmic domain.

Figure 2

Tertiary structures of the extracellular domains of CD58 (A) and CD2 (B). The structures were predicted by homology modeling with human CD58 (PDB 1CCZ) and CD2 (PDB 1HNF) extracellular domains as templates, showing the overall homology with human counterparts. The β strands of N-terminal IgV-like domains (AGFCC′C″BED) and the membrane-proximal IgC2-like domains (AEBDCFG and AEBCFG) of CD58 (blue) and CD2 (yellow) are labeled. (C) Detailed views of anti-parallel β-strands in stacked pleated β-sheets, α-helix structures, hydrophobic loops, Cys residues, and residues across the CD2–CD58 interface are depicted and labeled.

Figure 3

In vivo knockdown assay of cd58. (A) Screening of the most effective small interfering RNA (siRNA) interfering zebrafish cd58 expression. Three designed siRNAs encoding DNA oligonucleotides targeting different regions of each zebrafish cd58 mRNA were constructed to pSUPER vectors (cd58siRNA-1–3), which were in turn co-transfected with overexpression pcDNA6-cd58 plasmid into HEK293T cells, respectively. Inhibitory efficiencies were measured by real-time PCR. (B) After choosing the most effective siRNA targeting CD58, an LV harboring the siRNA was produced. The infection ability of the constructed LV was evaluated by detecting the GFP fluorescence release in HEK293T cells under a fluorescent microscope (Zeiss Axiovert 40 CFL; Zeiss, Jena, Germany) with original magnification at 400×. Scale bars, 200 µm. (C) The interfering abilities of the LV were examined in HEK293T cells transfected with the target gene overexpression plasmid in vitro by real-time PCR analyses. (D) The LV titer was assessed by the percentage of GFP⁺ HEK293T cells after exposure to different dilutions of LVs using flow cytometry (FCM). The gray histograms show background fluorescence on control cells without LV infection. The numbers above the bracketed lines indicate the percentage of cells in each. Data are from three independent experiments. (E–G) Detection of the inhibitory effect of CD58siRNA-LV by in vivo administration through real-time PCR and FCM analyses. The expression of cd58 in peripheral blood leukocytes (PBLs) and head kidney lymphocytes (HKLs) was detected using real-time PCR (E,F). For FCM analysis, the scrambled siRNA-LV (red) or cd58siRNA-LV (green) treated leukocytes was labeled with FITC-anti-Cd58. Scrambled siRNA-LV was set as control. Mean ± SE of results from three independent experiments are shown (*p < 0.05, **p < 0.01).

Figure 4

Subcellular localization analyses of Cd58 (B) and Cd2 (C) in HEK293T cells. Cells were transfected with empty control (A) or pEGFP-cd58 and pEGFP-cd2 (B,C) plasmids. Cd58 and Cd2 were clearly located on the cell surface membranes. Images were captured under a two-photon laser scanning confocal microscope (Zeiss LSM-710; original magnification, 630×). Scale bars, 10 µm.

Figure 5

Tissue and cellular distribution analyses of Cd58 and Cd2. (A,B) Tissue distribution analyses of cd58 (A) and cd2 (B) in control fish or in antigen [keyhole limpet hemocyanin (KLH) or Aeromonas hydrophila] challenged fish by real-time PCR. The relative gene expression profiles of cd58 and cd2 in various tissues were displayed as relative values to the expression levels of β-actin (cd58/β-actin and cd2/β-actin), respectively. Each sample was obtained from 10 fish and run in triplicate parallel reactions. The experiments were repeated independently at least three times (*p < 0.05, **p < 0.01). (C,D) Immunofluorescence staining of the leukocytes separated from the blood, spleen, and kidney tissues of the fish stimulated with KLH (in combination with lipopolysaccharide). Cells were stained with rabbit anti-Cd58 together with mouse anti-MHC class II (Mhc-ii), mouse anti-Cd80/86, or mouse anti-Cd83 (C), or stained with mouse anti-Cd2 together with rabbit anti-Cd4, rabbit anti-Tcr-α, or rabbit anti-Tcr-β (D). Nonrelated mouse and rabbit IgG isotypes were used as negative controls (data not shown). DAPI stain shows the location of the nuclei. The images were obtained using a two-photon laser scanning confocal microscope (Zeiss LSM-710; original magnification, 630×). An enlarged image of the target cell was generated by fourfold magnification. Scale bars, 2 μm.

Figure 6

Induced expression analysis of cd58 and cd2 on APCs or Cd4⁺ T cells upon keyhole limpet hemocyanin (KLH) [plus lipopolysaccharide (LPS)] or Aeromonas hydrophila stimulations. (A,B) Flow cytometric analysis of Cd58 or Cd2 expression level on APCs or Cd4⁺ T cells, which were sorted from peripheral blood, spleen, and kidney tissues 3 days after i.p. stimulation with PBS, KLH, LPS, KLH plus LPS, or A. hydrophila. The numbers above the outlined areas indicate the percentage of double positive cells in each group. (C) Flow cytometric analysis of Cd58⁺Mhc-ii⁺ cells from sorted APCs upon pulsing with PBS, KLH, LPS, KLH plus LPS, or A. hydrophila for 8 h in vitro. The numbers above the outlined areas indicate the percentage of double positive cells in each group. Means ± SE of results from three independent experiments are shown (*p < 0.05, **p < 0.01). (D) Real-time PCR analysis for the expression of Cd58 in APCs of each in vivo treatment group. (E) Real-time PCR analysis for the expression of cd58 in APCs with in vitro treatment of PBS, KLH, LPS, KLH plus LPS, or A. hydrophila. (F) Real-time PCR analysis for the expression of cd2 in Cd4⁺ T cells of each in vivo treatment group. The relative expression values were averaged from the data in three parallel reactions, and the results were obtained from at least three independent experiments (*p < 0.05, **p < 0.01).

Figure 7

In vitro evaluation of Cd58 and Cd2 in APC-initiated Cd4⁺ T cell proliferation. The proliferation of Cd4⁺ T_KLH (A) or Cd4⁺ T_A.h (B) was inhibited by treating the cells with anti-Cd58 or anti-Cd2 antibodies, determined by CFSE dilution through flow cytometry and by the expression levels of lck and cd154 (C) and cytokines (il-4/13a, il-4/13b, il-2, and ifn-γ) production (D) through real-time PCR. Cd4⁺ T_KLH cells or Cd4⁺ T_A.h co-cultured with PBS-loaded primary APCs were used as control. Error bars represent SE. All data were from at least three independent experiments (*p < 0.05, **p < 0.01).

Figure 7

In vitro evaluation of Cd58 and Cd2 in APC-initiated Cd4⁺ T cell proliferation. The proliferation of Cd4⁺ T_KLH (A) or Cd4⁺ T_A.h (B) was inhibited by treating the cells with anti-Cd58 or anti-Cd2 antibodies, determined by CFSE dilution through flow cytometry and by the expression levels of lck and cd154 (C) and cytokines (il-4/13a, il-4/13b, il-2, and ifn-γ) production (D) through real-time PCR. Cd4⁺ T_KLH cells or Cd4⁺ T_A.h co-cultured with PBS-loaded primary APCs were used as control. Error bars represent SE. All data were from at least three independent experiments (*p < 0.05, **p < 0.01).

Figure 8

In vivo evaluation of Cd58 in Cd4⁺ T cell activation. The degree of Cd4⁺ T cell activation is represented by the percentage of Cd4⁺Cd154⁺ T cells determined by flow cytometry (A,B) and by the expression levels of lck and cd154 genes detected by real-time PCR (C,D). In the flow cytometric analysis, different treatments were presented at the top of each block diagram. The numbers adjacent to the outlined areas indicate the percentage of Cd4⁺Cd154⁺ cells in each treatment group. In the real-time PCR assay, PCRs were run in combination with the endogenous β-actin control. Error bars represent SE. All data are from at least three independent experiments (*p < 0.05, **p < 0.01).

Figure 9

In vivo evaluation of Cd58 and Cd2 in B cell activation, antibody production and vaccinated immunoprotection. (A) Involvement of Cd58 and Cd2 in B cell activation. The degree of B cell activation is represented by the percentage of mIgM⁺Cd40⁺ cells determined by flow cytometry. The experimental treatments are presented at the top of each block diagram. The data above the outlined area in each block diagram indicate the average percentage of mIgM⁺Cd40⁺ B cells in each treatment group. (B) Involvement of Cd58 and Cd2 in IgM production. The titer of IgM against keyhole limpet hemocyanin (KLH) in each treatment group was examined by ELISA (n = 150) (*p < 0.05, **p < 0.01). (C) Blockade of Cd58 impairs the vaccinated immunoprotection against bacterial (Aeromonas hydrophila) challenge. Data points are from three independent experiments (n = 30); Differences were analyzed using log-rank test (**p < 0.01).

Figure 10

Functional evaluation of Cd58 and Cd2 interaction by introducing a recombinant soluble Cd2 protein (sCd2). (A). The degree of Cd4⁺ T cell activation was represented by the percentage of Cd4⁺Cd154⁺ cells determined by flow cytometry (FCM). The experimental treatments were presented at the top of each block diagram. The data above the outlined area in each block diagram indicated the average percentage of Cd4⁺Cd154⁺ T cells in each treatment group. (B) The degree of B cell activation was represented by the percentage of mIgM⁺Cd40⁺ cells determined by FCM. The experimental treatments were presented at the top of each block diagram. The data above the outlined area in each block diagram indicated the average percentage of mIgM⁺Cd40⁺ B cells in each treatment group. Fishes i.p. injected with MBP were used as control. Data points were from three independent experiments (n = 30).